Problem Set 1

Part A

1. This command counts the total number of rows for Region 3

cut -d ',' -f 2 recs2015_public_v3.csv | grep 3 | wc -l

2. This command takes the column variables: DOEID, NWEIGHT, andBRRWT1-BRRWT96 and saves the data into a compressed file named ‘newData’

cut -d ',' -f 1,475-571 recs2015_public_v3.csv | gzip > newData.gz

Part B

1. This is a Bash for loop that counts and prints the number of observations within each region

for region in 1 2 3 4
do

  echo $region
  cut -d ',' -f 2 recs2015_public_v3.csv| grep $region | wc -l

done

2. This command produces a file named region_division.txt and provides a sorted list showing unique combinations of values from REGIONC and DIVISION.

cut -d ',' -f 2,3 recs2015_public_v3.csv| sort -n | uniq > region_division.txt

Part A

In this problem we show which airlines were responsible for at least 1% of the flights departing any of the three NYC airports between January 1 and October 31, 2013. The results are shown in the following tables.

Part B

Here we compare the number and percent of annual flights using the airlines from Part A, but in year 2014 instead.

The Flights with the Largest Percent Increase and Decrease from 2013 to 2014 are shown on the table below.

We notice that some airlines show an increase in the percent of flights, even though their number of flights decreased. Take United Airlines Inc for example. They had 48,880 flights in 2013 and 46,267 flights in 2014. The main reason being that 2014 had less total flights than the total flights in 2013.

Part C

Now we show the percent flights of all three airports for each airline from Part A. Each table summary of the airports is followed by another table showing the carrier at that airport with the largest percent of flights.

Part A

Using the RECS 2015 data, first we compute the percent of homes that have stucco construction as their major outside wall material within each division. We compute (RSEs) using the procedure provided in the RECS 2015 documentation.

Part B

Now we show average total electricity usage in kilowatt hours per division. We illustrate our results in the table below and also graphically with a bar chart.

From the bar chart above, East South Central has the highest average total electricity usage. Next, we would like to compute average total electricity usage stratified by urban and rural status.

Part C

Here we show percents of Internet Users by Division and by Urban and Rural Status. We illustrate our results in the table below and also graphically with a bar chart. The goal is to identify the division with the largest disparity between Urban and Rural percentages of Internet Users.

In the table below, we show the division with the largest absolute difference between Urban and Rural percentage. Mountain South has the largest disparity between percent of Urban and Rural Internet Users.